Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method of classifying data of virtual machines in a heterogeneous computing environment that includes virtual machines and non-virtual machines, wherein the method is performed by one or more computing systems, each computing system having a processor and memory, the method comprising: accessing, by the one or more computing systems, at least some data stored by a virtual machine; classifying the at least some data to create classified data of the virtual machine, wherein classifying the at least some data of the virtual machine includes creating metadata associated with the classified data of the virtual machine; storing at least the metadata in an index, wherein the index represents, at least in part, the classified data of the virtual machine, and wherein the index also represents, at least in part, classified data of at least one non-virtual machine; detecting changes to the at least some data stored by the virtual machine, wherein the detecting changes to the at least some data stored on the virtual machine includes accessing a journal file for tracking operations performed on the at least some data stored on the virtual machine; updating the index in accordance with the detected changes; storing the updated index; determining distinct data objects within the at least some data stored by the virtual machine; indexing, the distinct data objects; and storing the indexed distinct data objects in a second index.
2. The method of claim 1 , wherein the virtual machine stores data in a file residing on a filesystem of a virtual machine host hosting the virtual machine, wherein detecting changes to the at least some data stored by the virtual machine includes detecting a modification to the file, and wherein the method further comprises: accessing the file, wherein the file includes multiple data objects and wherein at least one of the multiple data objects has been modified; determining the at least one data object that has been modified; accessing or creating metadata associated with the at least one data object in accordance with the modification; updating the index with the accessed or created metadata associated with the at least one data object; and storing the updated index.
This invention relates to virtual machine data management, specifically tracking changes to data stored by a virtual machine (VM) in a host filesystem. The problem addressed is efficiently detecting and indexing modifications to VM-stored data to maintain accurate metadata for monitoring, analysis, or backup purposes. The method involves a VM storing data in a file on the host filesystem. When the VM modifies data, the system detects changes by monitoring the file. The file contains multiple data objects, and the system identifies which specific objects have been modified. For each modified object, the system accesses or generates metadata describing the change, such as timestamps, content hashes, or other attributes. This metadata is then used to update an index, which serves as a centralized record of modifications. The updated index is stored for future reference, enabling efficient tracking of data changes over time. This approach ensures that metadata remains synchronized with the actual data state, improving reliability for applications that depend on accurate change detection, such as backup systems or data integrity verification tools. The method avoids the need for VM-internal monitoring, instead leveraging host-level filesystem observations to streamline the process.
3. The method of claim 1 , further comprising: for at least one of the distinct data objects: generating a substantially unique identifier for the distinct data object; determining, based on the substantially unique identifier, if an instance of the distinct data object has already been stored on a single instance storage device; and if an instance of the distinct data object has not already been stored on the single instance storage device, then storing the data object on the single instance storage device.
This invention relates to data storage systems that prevent redundant storage of identical data objects. The problem addressed is inefficient storage usage due to duplicate copies of the same data being stored multiple times, which wastes storage capacity and increases costs. The solution involves generating a unique identifier for each data object, checking whether an identical copy already exists in storage, and only storing the data object if no duplicate is found. This ensures that each distinct data object is stored only once, reducing redundancy and optimizing storage efficiency. The system works by first generating a substantially unique identifier for each data object, which can be derived from the data's content or other attributes. The identifier is then used to determine if an identical instance of the data object already exists in a single instance storage device. If no duplicate is found, the data object is stored. This approach is particularly useful in large-scale storage systems where minimizing redundancy is critical. The method ensures that storage resources are used efficiently by eliminating duplicate copies of the same data, thereby reducing storage costs and improving system performance. The invention can be applied in various storage environments, including cloud storage, databases, and file systems, to enhance storage efficiency and reduce redundancy.
4. The method of claim 1 further comprising: automatically determining multiple virtual machines hosted by a virtual machine host; determining a threshold number of simultaneous storage operations that can be performed upon data of the multiple virtual machines hosted by the virtual machine host; and performing multiple storage operations upon the data of the multiple virtual machines hosted by the virtual machine host, wherein the number of simultaneous copy operations performed upon the data of the multiple virtual machines hosted by the virtual machine host does not exceed the threshold number.
This invention relates to virtual machine (VM) storage management in cloud or data center environments. The problem addressed is inefficient or excessive storage operations in virtualized systems, which can lead to performance degradation, resource contention, or system instability when multiple VMs hosted on a single physical machine perform concurrent storage operations. The solution involves a method for managing storage operations across multiple VMs hosted by a single virtual machine host. The system first identifies all VMs running on the host. It then determines a threshold limit for the number of simultaneous storage operations that can be safely performed on the VMs' data without overloading the host's resources. This threshold is dynamically calculated based on the host's available capacity, performance metrics, or predefined policies. The system then executes storage operations (such as read, write, or copy operations) across the VMs while ensuring the total number of concurrent operations does not exceed the threshold. This prevents resource exhaustion and maintains system stability. The method may also prioritize operations based on urgency or VM importance. The approach ensures efficient storage management while avoiding performance bottlenecks in virtualized environments.
5. The method of claim 1 further comprising monitoring for changes to the at least some data stored by the virtual machine.
A system and method for managing data in a virtualized computing environment addresses the challenge of efficiently tracking and updating data stored by virtual machines (VMs) to ensure consistency and reliability. The invention involves a process where a virtual machine executes a workload that generates or modifies data, which is then stored in a storage system. The system monitors the stored data for changes, allowing for real-time or periodic detection of modifications. This monitoring can be performed by the virtual machine itself or by an external monitoring component that tracks data access patterns, file modifications, or metadata updates. The detected changes are then logged or reported to a management system, which can trigger further actions such as data synchronization, backups, or notifications. The method ensures that data integrity is maintained across distributed or virtualized environments, particularly in scenarios where multiple VMs may access or modify the same data. The solution is applicable in cloud computing, enterprise virtualization, and distributed storage systems where data consistency and tracking are critical.
6. At least one non-transitory computer-readable medium carrying instructions, which when executed by at least one data processor, executes operations to classify data of virtual machines in a heterogeneous computing environment that includes virtual machines and non-virtual machines, the operations comprising: accessing, by one or more computing systems, at least some data stored by a virtual machine; classifying the at least some data to create classified data of the virtual machine, wherein classifying the at least some data of the virtual machine includes accessing or creating metadata associated with the classified data of the virtual machine; storing at least the metadata in an index, wherein the index represents, at least in part, the classified data of the virtual machine, and wherein the index also represents, at least in part, classified data of at least one non-virtual machine; detecting changes to the at least some data stored by the virtual machine, wherein detecting changes to the at least some data stored on the virtual machine includes accessing a journal file for tracking operations performed on the at least some data stored on the virtual machine; accessing or creating metadata in accordance with the detected changes; updating the index with the accessed or created metadata; storing the updated index; determining distinct data objects within the at least some data stored by the virtual machine; indexing, the distinct data objects; and storing the indexed distinct data objects in a second index.
This invention relates to data classification and indexing in heterogeneous computing environments that include both virtual machines (VMs) and non-virtual machines. The problem addressed is the challenge of efficiently classifying and indexing data across diverse computing systems, particularly in environments where virtualized and non-virtualized systems coexist. The solution involves a computer-readable medium carrying instructions for executing operations to classify and index data from virtual machines while integrating this information with data from non-virtual machines. The system accesses data stored by a virtual machine and classifies it, generating metadata associated with the classified data. This metadata is stored in an index that represents both the virtual machine's data and data from at least one non-virtual machine. The system detects changes to the virtual machine's data by accessing a journal file that tracks operations performed on the data. When changes are detected, the system updates the metadata and the index accordingly. Additionally, the system identifies distinct data objects within the virtual machine's data, indexes these objects, and stores them in a separate index. This approach ensures that data from virtual and non-virtual machines is consistently classified, indexed, and updated, facilitating efficient data management in heterogeneous environments.
7. The non-transitory medium of claim 6 , wherein the virtual machine stores data in a file residing on a filesystem of a virtual machine host hosting the virtual machine, wherein detecting changes to the at least some data stored by the virtual machine includes detecting a modification to the file, and wherein the method further comprises: accessing the file, wherein the file includes multiple data objects and wherein at least one of the multiple data objects has been modified; determining the at least one data object that has been modified; accessing or creating metadata associated with the at least one data object in accordance with the modification; updating the index with the accessed or created metadata associated with the at least one data object; and storing the updated index.
A system and method for tracking data modifications in a virtualized environment involves monitoring changes to data stored by a virtual machine (VM) on a host filesystem. The VM writes data to a file on the host's filesystem, and the system detects modifications to this file to identify changes in the VM's data. The file contains multiple data objects, and when a modification occurs, the system determines which specific data objects have been altered. Metadata associated with these modified data objects is then accessed or created, and the system updates an index with this metadata. The updated index is stored to maintain a record of the changes. This approach enables efficient tracking of data modifications within a VM by leveraging filesystem-level monitoring and metadata management, ensuring accurate and up-to-date indexing of modified data objects. The system is particularly useful in environments where VM data integrity and change tracking are critical, such as in cloud computing or virtualized storage systems.
8. The non-transitory medium of claim 6 , further comprising: for at least one of the distinct data objects: generating a substantially unique identifier for the distinct data object; determining, based on the substantially unique identifier, if an instance of the distinct data object has already been stored on a single instance storage device; and if an instance of the distinct data object has not already been stored on the single instance storage device, then storing the data object on the single instance storage device.
This invention relates to data storage systems, specifically addressing the challenge of efficiently managing and storing distinct data objects to avoid redundancy. The system generates a substantially unique identifier for each distinct data object to determine whether an instance of that object already exists in a single instance storage device. If the object is not already stored, it is added to the storage device. This process ensures that only one copy of each distinct data object is maintained, reducing storage redundancy and optimizing storage space. The system may also include mechanisms for handling metadata associated with the data objects, such as tracking relationships between objects or managing access permissions. The unique identifier generation and storage verification steps are applied to each distinct data object to maintain consistency and prevent duplicate storage. This approach is particularly useful in environments where large volumes of data are processed, such as databases, cloud storage, or distributed systems, where minimizing redundant storage is critical for performance and cost efficiency.
9. The non-transitory medium of claim 6 , further comprising automatically determining multiple virtual machines hosted by a virtual machine host; determining a threshold number of simultaneous storage operations that can be performed upon data of the multiple virtual machines hosted by the virtual machine host; and performing multiple storage operations upon the data of the multiple virtual machines hosted by the virtual machine host, wherein the number of simultaneous copy operations performed upon the data of the multiple virtual machines hosted by the virtual machine host does not exceed the threshold number.
In the field of virtualized computing environments, managing storage operations for multiple virtual machines (VMs) hosted on a single virtual machine host can lead to performance bottlenecks and resource contention. When storage operations such as data copying are performed simultaneously across multiple VMs, excessive concurrent operations can degrade system performance, overwhelm storage resources, or violate service-level agreements. This invention addresses the problem by providing a method to optimize storage operations in a virtualized environment. The system automatically identifies all virtual machines hosted by a virtual machine host and determines a threshold number of simultaneous storage operations that can be performed on the data of these VMs without exceeding system capacity or performance limits. The system then executes multiple storage operations, such as copying data, across the VMs while ensuring that the number of concurrent operations does not exceed the predefined threshold. This controlled approach prevents resource overload while maintaining efficient data management. The solution is particularly useful in cloud computing and enterprise virtualization environments where multiple VMs share storage resources.
10. The non-transitory medium of claim 6 , further comprising monitoring for changes to the at least some data stored by the virtual machine.
A system and method for managing data in a virtualized computing environment addresses the challenge of efficiently tracking and responding to changes in data stored by virtual machines. The system includes a virtual machine configured to execute applications and store data, and a monitoring module that detects modifications to at least some of the stored data. The monitoring module operates in real-time or periodically to identify changes, which may include additions, deletions, or updates to files, databases, or other data structures. The system may also include a synchronization module that propagates detected changes to other systems or storage locations, ensuring data consistency across distributed environments. Additionally, the system may implement access control mechanisms to restrict or log data modifications, enhancing security and auditability. The monitoring module may use checksums, timestamps, or file system hooks to track changes, while the synchronization module may employ protocols like rsync or custom APIs to transfer updates. This approach improves data integrity, reduces synchronization delays, and supports compliance with regulatory requirements by providing a comprehensive record of data modifications. The system is particularly useful in cloud computing, enterprise IT, and distributed storage applications where real-time data consistency is critical.
11. The non-transitory medium of claim 6 , wherein updating the index is performed by a journaling agent, the journaling agent includes a virtual filter driver module.
A system and method for managing file system operations involves a journaling agent that updates an index of file system metadata. The journaling agent includes a virtual filter driver module that intercepts and processes file system operations, such as reads, writes, and deletions, before they are committed to the file system. The virtual filter driver module monitors these operations and records relevant metadata, such as file attributes, timestamps, and access patterns, in a structured index. This index is used to track changes to the file system in real-time, enabling efficient data recovery, auditing, and analysis. The journaling agent ensures that the index is updated consistently and accurately, even in the event of system failures or interruptions. The system is designed to operate transparently, without requiring modifications to existing applications or file system structures. This approach improves data integrity, simplifies backup and recovery processes, and provides a comprehensive record of file system activity for security and compliance purposes. The solution is particularly useful in environments where real-time monitoring and historical tracking of file operations are critical, such as enterprise systems, cloud storage, and regulated industries.
12. The non-transitory medium of claim 11 , wherein the journaling agent is deployed on a virtual I/O port or data stack.
A system for managing data storage operations in a virtualized computing environment addresses the challenge of ensuring data integrity and consistency during input/output (I/O) operations in virtualized systems. The system includes a journaling agent that captures and records changes to data before they are written to a storage device, creating a log of modifications. This log allows for recovery of data in case of system failures or disruptions, ensuring that data remains consistent and intact. The journaling agent is deployed on a virtual I/O port or data stack, enabling it to intercept and monitor I/O operations at a low level within the virtualized infrastructure. By operating at this level, the journaling agent can efficiently track changes without significant performance overhead, while also providing a centralized point for managing data integrity across multiple virtual machines or storage devices. The system may also include a recovery module that uses the journal to restore data to a consistent state following a failure, ensuring minimal data loss and downtime. This approach is particularly useful in environments where high availability and data reliability are critical, such as enterprise storage systems or cloud computing platforms.
13. The non-transitory medium of claim 12 , wherein the journaling agent operates in conjunction with a virtual file management module to record operations executed on the virtual data.
A system for managing virtual data operations includes a journaling agent that records operations performed on virtual data. The journaling agent operates in conjunction with a virtual file management module to track and log these operations. The virtual file management module handles the creation, modification, and deletion of virtual data, while the journaling agent ensures that all operations are recorded in a journal for later reference or recovery. This system is designed to enhance data integrity and reliability in virtual environments by maintaining a detailed log of all actions performed on virtual data. The journaling agent may also include features to optimize storage and retrieval of the recorded operations, ensuring efficient access to the journal data when needed. The system is particularly useful in environments where data consistency and recoverability are critical, such as in virtualized storage systems or cloud computing platforms. By integrating the journaling agent with the virtual file management module, the system provides a comprehensive solution for tracking and managing virtual data operations, ensuring that all changes are documented and can be audited or reverted if necessary.
14. A system for classifying data of virtual machines in a heterogeneous computing environment that includes virtual machines and non-virtual machines the system comprising: one or more computing systems, each computing system having a processor and memory, the one or more computing systems configured to: access at least some data stored by a virtual machine; classify the at least some data to create classified data of the virtual machine; store at least the metadata in an index, wherein the index represents, at least in part, the classified data of the virtual machine, and at least in part, classified data of at least one non-virtual machine; detect changes to the at least some data stored by the virtual machine, wherein the detecting changes to the at least some data stored on the virtual machine includes accessing a journal file that tracks operations performed on the at least some data stored on the virtual machine; update the index in accordance with the detected changes; store the updated index; determine distinct data objects within the at least some data stored by the virtual machine; index the distinct data objects; and store the indexed distinct data objects in a second index.
The system classifies and indexes data from virtual machines (VMs) and non-virtual machines in a heterogeneous computing environment. The environment includes both VMs and traditional computing systems, where data classification and indexing are essential for efficient management and retrieval. The system addresses challenges in tracking and organizing data across diverse computing resources, ensuring consistency and accessibility. The system uses one or more computing systems, each with a processor and memory, to perform the classification and indexing tasks. It accesses data stored by a VM, classifies the data, and stores metadata in an index. This index represents classified data from both VMs and non-virtual machines, enabling unified data management. The system detects changes to VM data by monitoring a journal file that tracks operations performed on the data, ensuring the index remains up-to-date. When changes are detected, the index is updated and stored. Additionally, the system identifies distinct data objects within the VM's data, indexes these objects, and stores them in a separate index. This allows for granular data tracking and retrieval. The system ensures that data from VMs and non-virtual machines is consistently classified, indexed, and updated, improving data management in heterogeneous environments.
Unknown
August 25, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.